{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:PKOKYXHAVVXWRLNWMG4UNOMRMP","short_pith_number":"pith:PKOKYXHA","schema_version":"1.0","canonical_sha256":"7a9cac5ce0ad6f68adb661b946b99163e82cf18833aebf6a3aaaffb574c66a31","source":{"kind":"arxiv","id":"2410.22391","version":3},"attestation_state":"computed","paper":{"title":"A Large Recurrent Action Model: xLSTM enables Fast Inference for Robotics Tasks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"G\\\"unter Klambauer, Johannes Brandstetter, Korbinian P\\\"oppel, Maximilian Beck, Razvan Pascanu, Sepp Hochreiter, Thomas Adler, Thomas Schmied, Vihang Patil","submitted_at":"2024-10-29T17:55:47Z","abstract_excerpt":"In recent years, there has been a trend in the field of Reinforcement Learning (RL) towards large action models trained offline on large-scale datasets via sequence modeling. Existing models are primarily based on the Transformer architecture, which result in powerful agents. However, due to slow inference times, Transformer-based approaches are impractical for real-time applications, such as robotics. Recently, modern recurrent architectures, such as xLSTM and Mamba, have been proposed that exhibit parallelization benefits during training similar to the Transformer architecture while offering"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2410.22391","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-10-29T17:55:47Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f48ca68976865e83a6249566c6238a0376333000524c910e558928b589f3fa28","abstract_canon_sha256":"497c6a1a95099149d4e2fbf4d09f2ea80801a6e3057697576a9b9c14d89761eb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:16:04.272690Z","signature_b64":"FpDWfO3qEefuEU39b+WzK9qmUn+/M+KQh57jSEO2fl8i5zCwWTx33PM7hS9K/+ACcMDdpepJPKKp3pxsjo+qCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7a9cac5ce0ad6f68adb661b946b99163e82cf18833aebf6a3aaaffb574c66a31","last_reissued_at":"2026-07-05T11:16:04.272164Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:16:04.272164Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Large Recurrent Action Model: xLSTM enables Fast Inference for Robotics Tasks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"G\\\"unter Klambauer, Johannes Brandstetter, Korbinian P\\\"oppel, Maximilian Beck, Razvan Pascanu, Sepp Hochreiter, Thomas Adler, Thomas Schmied, Vihang Patil","submitted_at":"2024-10-29T17:55:47Z","abstract_excerpt":"In recent years, there has been a trend in the field of Reinforcement Learning (RL) towards large action models trained offline on large-scale datasets via sequence modeling. Existing models are primarily based on the Transformer architecture, which result in powerful agents. However, due to slow inference times, Transformer-based approaches are impractical for real-time applications, such as robotics. Recently, modern recurrent architectures, such as xLSTM and Mamba, have been proposed that exhibit parallelization benefits during training similar to the Transformer architecture while offering"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.22391","kind":"arxiv","version":3},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2410.22391/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2410.22391","created_at":"2026-07-05T11:16:04.272220+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.22391v3","created_at":"2026-07-05T11:16:04.272220+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.22391","created_at":"2026-07-05T11:16:04.272220+00:00"},{"alias_kind":"pith_short_12","alias_value":"PKOKYXHAVVXW","created_at":"2026-07-05T11:16:04.272220+00:00"},{"alias_kind":"pith_short_16","alias_value":"PKOKYXHAVVXWRLNW","created_at":"2026-07-05T11:16:04.272220+00:00"},{"alias_kind":"pith_short_8","alias_value":"PKOKYXHA","created_at":"2026-07-05T11:16:04.272220+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.11997","citing_title":"pLSTM: parallelizable Linear Source Transition Mark networks","ref_index":22,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PKOKYXHAVVXWRLNWMG4UNOMRMP","json":"https://pith.science/pith/PKOKYXHAVVXWRLNWMG4UNOMRMP.json","graph_json":"https://pith.science/api/pith-number/PKOKYXHAVVXWRLNWMG4UNOMRMP/graph.json","events_json":"https://pith.science/api/pith-number/PKOKYXHAVVXWRLNWMG4UNOMRMP/events.json","paper":"https://pith.science/paper/PKOKYXHA"},"agent_actions":{"view_html":"https://pith.science/pith/PKOKYXHAVVXWRLNWMG4UNOMRMP","download_json":"https://pith.science/pith/PKOKYXHAVVXWRLNWMG4UNOMRMP.json","view_paper":"https://pith.science/paper/PKOKYXHA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.22391&json=true","fetch_graph":"https://pith.science/api/pith-number/PKOKYXHAVVXWRLNWMG4UNOMRMP/graph.json","fetch_events":"https://pith.science/api/pith-number/PKOKYXHAVVXWRLNWMG4UNOMRMP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PKOKYXHAVVXWRLNWMG4UNOMRMP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PKOKYXHAVVXWRLNWMG4UNOMRMP/action/storage_attestation","attest_author":"https://pith.science/pith/PKOKYXHAVVXWRLNWMG4UNOMRMP/action/author_attestation","sign_citation":"https://pith.science/pith/PKOKYXHAVVXWRLNWMG4UNOMRMP/action/citation_signature","submit_replication":"https://pith.science/pith/PKOKYXHAVVXWRLNWMG4UNOMRMP/action/replication_record"}},"created_at":"2026-07-05T11:16:04.272220+00:00","updated_at":"2026-07-05T11:16:04.272220+00:00"}